A Chaotic Dynamic Congestion Prediction Method under the Space-Terrestrial Integrated Network

Abstract

The space-terrestrial integrated network is an important research field of the future network. The dynamic and heterogeneous nature of the space-terrestrial integrated network brings challenges to the research of congestion prediction methods. In this paper, we propose a chaotic dynamic congestion prediction method under the space-terrestrial integrated network, which solves the problems of insufficient dynamics and low accuracy of the existing congestion prediction methods for the space-terrestrial integrated network. The chaotic dynamic congestion prediction method is to predict the time series by using the GRU neural network model of the improved particle swarm algorithm to optimize the parameters after wavelet analysis. The experimental results show that the prediction accuracy of the chaotic dynamic congestion prediction method is higher, and it is more suitable for the space-terrestrial integrated network.